DE2653837C2 - Scintillator arrangement of a scintillation camera - Google Patents

Description

20th

30th

35

The invention relates to a scintillator arrangement of a scintillation camera in the preamble of Claim specified genus.

Such a scintillator arrangement is from "IEEE Transactions on Nuclear Science", Volume NS-22, February 1975, pages 626 to 632. The ones there in particular Scintillator arrangements used as X-ray detectors work with scintillation crystals, which are used in Smaller diameters have a thickness of 3 mm, for large-area detectors a thickness of 5 mm. A small thickness is desirable there in order to keep the volume of the scintillation crystal as small as possible, and thus as small as possible to keep the background noise correspondingly low.

With scintillation cameras it is advantageous if on those photomultipliers that are further away from the luminescent site in the scintillation crystal, one the smallest possible amount of light impinges in order not to impair the spatial resolution.

To explain this in more detail, reference is made to FIG. 1, which shows a conventional scintillator arrangement of a scintillation camera. The light comes from a scintillation point P, from which scintillation light emanates when a gamma quantum hits, mainly on the photomultiplier PMi ^ i, PMi and PM _{i +} \ (with i = integer). In Fig. 1, an optical conductor 1 and a scintillation window 2 are also shown. The scintillation crystal 3 consists for example of NaI (Tl).

An optimal light distribution function would be provided if the light only hit the above-mentioned photomultiplier units adjacent to the scintillation point P , but not the photomultiplier units PMi-2, PMi-z , etc., which are further away. Such a light distribution function cannot be achieved with the previous scintillator arrangements, so that the known scintillation cameras have a low spatial resolution.

The invention is based on the object of the scintillator arrangement of a scintillation camera to be developed further in such a way that a higher spatial resolution is achieved.

The inventive solution to this problem is given in the characterizing part of the claim. As will be explained in detail below, the use of a scintillation crystal, the thickness specified in the preamble of the claim of a maximum of 9 mm and at the same time the im Characteristic part of the patent claim has specified surface roughness, to a light distribution

function that results in a high spatial resolution

From the book by Kai Siegbahn »ALPHA-, BETA-

AND GAMMA-RAY SPECTROSCOPY «, 1965, Volume 1, Page 283, although it is known To provide scintillation crystals with a surface roughness that is in the characteristic part of the Claim specified range also falls there is also from an improvement in resolution Speech. However, since the publication relates to spectrometers

ίο it can only refer to the term "dissolution" be about energy dissolution, d. H. by the extent to which the weak radiant energies of the Spectrometer still to be recorded. In contrast, the invention, as set out above, is concerned with the spatial resolution, which is essential for a scintillation camera

The subject of the earlier German patent 26 33 950 is the same task as the the present invention is based; the task is there with the scintillator arrangement of a scintillation camera by two arranged one behind the other in the direction perpendicular to the light exit plane Scintillation crystals dissolved on their surfaces facing each other to form a scattering element are roughly polished. The total thickness of the two scintillation crystals there is preferably 10 to 15mm, i.e. slightly larger than the thickness specified in the preamble of the claim in the invention formed scintillator arrangement present crystal. The thickness of one of the two crystals can therefore in the case of the subject matter of the earlier patent in the case of the arrangement designed according to the invention assumed range, in the exemplary embodiment between approximately 0.5 and 6 mm.

Fundamentals and an embodiment of the invention are explained below with reference to the drawings explained in more detail. In the drawings shows

1, to which reference has already been made, a schematic representation of a conventional scintillator arrangement,

F i g. 2 shows an arrangement for explaining what has been carried out in connection with the present invention Investigations,

F i g. 3 is a graph showing the relationship between signal amplitudes and the surface roughness of the scintillation crystal,

F i g. 4A and 4B are graphs to illustrate the relationship between the Surface roughness and light distribution,

F i g. 5 a further diagram to explain the influence of the surface roughness,

F i g. 6 an embodiment,

F i g. 7 is a diagram for explaining the light distribution in the FIG. 6 illustrated embodiment and

F i g. 8 is a table of the figures with the exemplary embodiment according to FIG. 6 achieved resolution values.

In connection with the present invention, studies have been made to find out the conditions under which an optimal light distribution function occurs; that is, it was investigated
i) the condition in which the gradient of the light distribution is greatest at the photomultiplier located closest to the luminescence point and the amount of light falling on it is small,
the condition under which the amount of light in the photomultiplier can be made zero, the one closest to that of the luminescent site

55

65

lying photomultiplier lie or arranged further away from the luminescence point are and

in) the conditions under which the number N of the total released photoelectrons can be increased.

In these experiments the curves for the light distribution for different crystal thicknesses were made using measured on a rough polished surface. These measurements showed a previously unknown, very interesting phenomenon to be described in detail below.

At first, attention was paid to an experiment to increase the total photoelectron number TV by rough polishing the crystal, the values shown in FIG. 2 was used. The thickness d and the surface roughness of the scintillation crystal 3 constituted the parameters. The diameter of the secondary electron multiplier PM and the thickness do of a glass window 2 were kept constant. 57 Co was used as a source delivering radiation γ . The measured values obtained are shown in FIG. 3, the pulse heights being plotted on the ordinate and the surface roughness on the abscissa.

From Fig. 1 it can be seen that the roughness, which corresponds to an average polishing agent grain diameter of 20 μm, results in a significantly greater pulse height for each crystal thickness than a surface that is mirror-polished. That is, the number N of total photoelectrons increases as the surface becomes rougher. The number N , on the other hand, is inversely proportional to the spatial resolution.

With a surface roughness below a value corresponding to an average polishing agent grain diameter of 40 μm, there was no significant increase in the number H for a crystal with a thickness of 12.7 mm, as shown in FIG. 3 by a curve marked with small circles is. If the crystal is thinner than 9 mm, however, the result is a number «o N, which increases with rougher surface roughness, as shown by the curve with small crosses. The rationale for this phenomenon is given below. A curve denoted by small triangles is shown in FIG. 3, these triangles being the measurement results obtained with a 5 mm thick crystal.

It has been established by different groups of crystals and by different polishing agents and polishing processes that for crystals thinner than 9 mm> °, the pulse height at grain diameters of the polishing agent over 40 μπί is essentially proportional to the grain diameter of the polishing agent, and that this tendency the clearer and more conspicuous it is, the thinner the crystals are. For crystals that are between 9 and 12.7 mm thick, however, the magnification of the number N was not as great.

In order to obtain an improved light distribution curve, i. i.e., to the terms and conditions in accordance with points i) and ii) listed above, then became an investigation performed to determine the relationship between the position of the luminescence point and the pulse height by changing the position of the gamma radiation source (see FIG. 2) in a lateral direction. the Results of this investigation are shown in FIGS. 4A and 4B. In these figures are the Light distribution functions shown in standardized form to facilitate evaluation. F i g. 4A shows Relative values of the pulse height for crystals with a thickness of 12.7 mm and a roughness accordingly mean polishing agent grain diameters of 24 μm (Measured values O), 74 μm (measured values Δ) and from 125 μm (Meas'verte x). Fig.4B shows the measurement results for a 5 mm thick crystal with roughness corresponding to average grain diameters of 24 μm (measured values O) and74 μπι (measured values Δ).

In FIG. 4A, the curves have a pronounced, sharp point and a trailing, slowly tapering end. Differences in roughness do not affect the tip of the curve, but the side end of the curve somewhat. Furthermore, this difference does not cause any significant change in the gradient in the position (Xi + 1) of the photomultiplier which is closest to the luminescence source

In FIG. 4B, the curves show a less slow tailing off of the curve end, i.e. h “the curves decrease more with increasing distance. The shape of the curve tip changes with greater roughness. The gradient at point Xi + 1 of the photomultiplier adjacent to the luminescence point rises sharply with increasing roughness. It should be noted that the pulse height is zero at the distance Xi + 2. This has been found generally for crystals thinner than 9 mm.

In FIGS. 4A and 4B, the relative values of the pulse heights are plotted on the ordinates and the distances from the scintillation point are plotted on the abscissa. The indications Xi, Xi + 1 and Xi + 2 designate the center points of the photomultiplier units PMi, PMi + 1 and PMi + 2, respectively.

It has thus been found that three requirements can be properly satisfied only by making the crystal thinner than 9 mm and rougher, the requirements being to increase the number N of photoelectrons, to increase the gradient, and to reduce the light at the photomultiplier located closest to the luminescent point as well as the light that strikes the photomultiplier units adjacent to it and on the remote photomultiplier units.

However, the question still remains as to whether this effect is actually due to the rougher surface or to the smaller overall distance d + do between the scintillation site and the photomultiplier. In order to find out this in the advantageous phenomenon described above, an experiment was carried out for crystals with thicknesses of 12.7 mm and 5.0 mm, in which the total distance d + do was kept constant at 18 mm by using suitable optical guides. The measurement results are shown in FIG. 5 shown. The information on the axes corresponds to that of FIG. 4, where the distances .Yin mm are given on the abscissa. The measured values χ give the curve for the 12.7 mm thick crystals, and the measured values give the curve for the 5.0 mm thick crystals which were polished with a polishing agent with a grain size of 74 μ, η. From Fig. 5 shows that the rough-polished crystal results in an ideal, trapezoidal course of the light distribution function FXx) regardless of the total distance between the luminescence point and the photomultiplier.

F i g. 6 shows the structure of a scintillator which, as a scintillation crystal, has a Nal crystal 5 with a thickness of 9 mm or less in thickness. A window 4 is made of glass, for example. A

The scattering reflector 6 consists for example of Al2O3, MgO or BaSO ₄ . The window 4 and the crystal 5 are optically coupled to one another with an epoxy resin adhesive. The upper and lower surfaces of the crystal 5 are roughly polished with a degree of polishing corresponding to an average abrasive grain diameter of 40 to 125 μm. The reflector 6 reflects that light back into the crystal which falls from the scintillation point on the underside of the crystal.

The optical processes in the crystal 5 should be based on FIG. 7 are described by Explain the reason why the rough polished surface shows a certain effect when the Crystal thickness is less than 9 mm.

7 shows two media with a higher refractive index π of 1.54 and a lower refractive index η of 1.0, which are roughly polished at the transition surfaces. When a laser beam is directed onto the media, it results in curves for the light emission as shown in FIG. 7 are shown. In Fig. 7 shows the light intensity (in a logarithmic representation) and the direction of radiation in polar coordinates, the points Oa, Ob and Oc being the poles or the zero points of the polar coordinate system. The direction of radiation or the direction of the light gives the angle of emission! of light again, which is the direction angle of the polar coordinate.

For the sake of easier understanding, it is assumed here that light with the same intensity in each case hits the points Oa. Ob and Oc occurs although the intensities are not the same in practice, as will be described below.

For the rougher transition surfaces is the "Oc component? _C at the point O in Lich teinfallsrichtung compared to the component Obb at the site Whether considerably smaller since the spread or dispersion with increasing incidence wink el is larger. In contrast, the Component e O _C C \ at the point O _c compared to the component Obb \ at the point Ob in the direction of the entrance surface of the photomultiplier is not so different, but if the surfaces are finely polished, as shown by the dashed curves, this tendency is just the opposite, that is, the decrease in the component in the direction of incidence or in the angle of incidence is small, and the decrease in the component which is directed onto the surface of the photomultiplier is relatively large.

In the case of thin crystals, the components with a larger angle of incidence are dominant, since the distance d from the scintillation point P to the polished surface is relatively small. Since the surface is rougher, the distribution curve of the emitted light becomes as shown in FIG. 7 is shown, also for this reason rounder, so that the light components directed onto the photomultiplier become larger. Or more specifically, since the distance between the scintillation point and each point is different from each other, the absolute value of the amount of light appearing at the points Oa, Ob and Oc must be calculated taking the square of the distance into account. However, the actual absolute values here are not significant enough to understand the advantages resulting from the smaller thickness of the crystal and the rough surface thereof.

From what has been described above, it follows that the effect of a rough polish on thin crystals of is of greater importance because the propagation characteristic depends heavily on the angle of incidence of light and because the Components of larger angles of incidence dominate in thin crystals. A 5 mm thick crystal shows therefore, a greater effect on different surface roughness than a 9 mm thick one Crystal.

In FIG. 4B, the change in the distribution around the top of the curve has been viewed as the result of a sharp rise or a large increase in the light component directed onto the photomultiplier due to the roughness of the crystal surface. Simultaneously, the steeply extending portion and the non trailing or slowly falling edge area are the Kurv en de r fact attributable to that the compo substantially a'als component falls NEN te Obb m ore g egenüber of component 0 ₃ OcCgegenüber component Obb so that the component in the direction of the angle of incidence decreases as the lateral distance increases.

In the case of a thick crystal, for reasons of comparison, the distance d ' from the scintillation point P' to the polished surface is selected to be relatively larger in order to create smaller angles of incidence. The straight line POc is therefore to be regarded as parallel to the straight line POb .

Therefore, the P 'coming and au ffalle to the point Oc walls Li CHT in the same manner as the light Obb' is widened and Obb, so that the vertical components (ie, a uf de n photomultiplier related components) Obb \ and Obb \ result.

As a result, in the case of thick crystals, when comparing a vertical incident fall and ey ies oblique incidence, the inequality O _a a \> Obb \ exists if the crystal is fe in polar . With rough- polished crystals the result is O _a a \> Obb \ and Obb \> Obb \. Therefore, the total light impinging on the photomultiplier is practically constant in the case of thick crystals and does not depend on the roughness of the crystal surface, since the integrated light components that are directed onto the photomultiplier are practically the same in both cases, so that the in F i g. 3 shows property shown. The trailing or only slowly sloping edge portions of the curves according to Figures 4A go back to the fact that a smaller angle of incidence with thick crystal causes all less decrease of the components Obb in the incident direction.

Cameras equipped with the described scintillator arrangement showed release values as shown in the table in FIG. 8 when ⁵⁷ Co is used as the radiation source.

As previously described, the benefits of a rough polish are greater with thinner crystals. It however, there are limits in practice with regard to the roughness and thickness of the crystals. A very thin crystal leads to a significant deterioration in the detection efficiency of gamma rays. Therefore, the crystal thickness for measuring the radiation of Tc (technetium) is at least 1 to 3 mm.

In addition, a very rough polish leads to the cracking of the crystal and the formation of Bubbles on the entire interface from the glass of the window to the crystal. When the crystal of combined domain areas is formed with different axes, there is also a different light propagation through the domains, so that an incorrect sensitivity distribution was measured will.

For this purpose 4 sheets of drawings

Claims (1)

Claim: Scintillator arrangement of a scintillation camera with a scintillation crystal with a thickness of 9 mm or less and means for optically coupling the scintillation crystal to the photomultiplier, characterized in that the surface facing the photo multipliers and that facing away from the photo multipliers Surface of the scintillation crystal (5) have a roughness that corresponds to a mean grain diameter of the polishing agent used to achieve it corresponds to 40 to 125 μπι